In a complementary metal oxide semiconductor (CMOS) image sensor called a “dual pixel,” a plurality of pixels (photo diodes) (hereinafter sometimes simply referred to as a “PD”) sharing an on-chip lens (hereinafter sometimes simply referred to as an “OCL”) shares the same floating diffusion (hereinafter sometimes simply referred to as an “FD”).
More specifically, the CMOS image sensor called the “dual pixel” has the structure of sharing an FD by two pixels. In such an image sensor, transfer gates (hereinafter sometimes simply referred to as “TRGs”) of two pixels are arranged adjacent to each other, and a polysilicon (Poly-Si) gate has a two-layer structure. With such a structure, narrowing of an FD region and expansion of a PD region can be realized, and as a result, sensitivity and a saturation signal amount (hereinafter sometimes referred to as a “Qs”) can be improved.